Recently, visual contents collected from surveillance cameras, mobile phones, personal photo collections, news footage, or medical images have been explosively increased. How to automatically/quantitatively analyze and understand the acquired visual contents is becoming one of the most active research areas in the vision community due to the scientifically challenging problems and its great benefits to real life applications. On the other hand, machine learning techniques especially the deep learning framework have manifested the surprising superiority for extracting structural and semantic visual representation in numerous computer vision applications such as image classification, object detection/localization, image segmentation, captioning, and so on.

With machine learning and computing techniques, it is prospected to discover the inherent structure of the available unconditioned visual contents and to achieve more promising results for different applications based on visual semantic analysis. This workshop calls for submissions on the emerging and challenging research topics on visual semantic analysis by exploiting advanced machine learning and computing techniques for wide applications such as surveillance, remote sensing, industrial defect detection, medical data analysis, and so on.